A system and method for wireless communications with codebook quantization are provided. A method for communications device operations includes estimating a channel between a controller and a mobile device, generating channel state information from the channel estimate, quantizing the channel state information by selecting a codeword from a rank n codebook based on a codeword selection mechanism, where n is a non-negative integer value, transmitting an index corresponding to the selected codeword and receiving a transmission from the controller. The transmission is precoded based on the index. The rank n codebook includes 2K codewords with each codeword comprising n columns of a transformed base matrix of dimension Nt×Nt, where K is a positive integer number representing a number of bits in the rank n codebook, and Nt is a number of transmit antennas at the controller.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for communications device operations, the method comprising: estimating, by the communications device, a channel between a controller and a mobile device; generating channel state information from the channel estimate; quantizing the channel state information by selecting a codeword from a rank n codebook in accordance with a codeword selection mechanism, where n is a non-negative integer value; transmitting an index corresponding to the selected codeword; and receiving a transmission from the controller, wherein the transmission is precoded in accordance with the index, wherein the rank n codebook comprises 2 K codewords with each codeword comprising n columns of a transformed base matrix of dimension N t ×N t , where K is a positive integer number representing a number of bits in the rank n codebook, and N t is a number of transmit antennas at the controller, and wherein an i-th codeword of the rank n codebook is expressible as C i =diag(u i )×DFT N t , where DFT N t is the base matrix, u i (n)=e jπ sin(k sin(θ i ))n , iε[0, . . . , 2 K −1], nε[0, . . . , N t −1], k is a constant, θ i ε[0,2π], and diag( ) is a function that returns diagonal elements of u i .
A method for wireless communication involves a mobile device estimating the communication channel to a base station. The mobile device generates channel state information and then quantizes this information by selecting the best matching codeword from a predefined "rank n" codebook. This codebook contains 2 to the power of K codewords. Each codeword is derived from a base matrix (Nt x Nt) transformed by a diagonal matrix. The mobile device then transmits the index of the selected codeword back to the base station. Finally, the mobile device receives data from the base station, which has been precoded using the received index. The i-th codeword (Ci) is calculated as diag(ui) multiplied by DFTNt, where ui(n) = e^(j * pi * sin(k * sin(theta_i)) * n). DFTNt is the base matrix.
2. The method of claim 1 , wherein the channel is estimated in accordance with a measurement of a pilot transmitted on the channel.
In the wireless communication method described above, the mobile device estimates the communication channel by measuring a pilot signal transmitted by the base station. This pilot signal provides the mobile device with information about the channel characteristics, allowing it to accurately estimate the channel conditions between the mobile device and the base station.
3. The method of claim 1 , wherein n is equal to one, and wherein k = 2 3 π 9 .
In the wireless communication method, the rank 'n' is set to 1, implying a single data stream. The constant 'k' used in calculating the codewords is set to a specific value: (2/3)*(pi/9). This fixed value of 'k' influences the characteristics of the codebook and the precoding process.
4. The method of claim 3 , wherein θ i = - π 3 + 2 i π 45 .
In the wireless communication method where rank n equals one, and k is (2/3)*(pi/9), the angle theta_i, used to generate codewords is calculated as theta_i = -pi/3 + 2 * i * pi/45. This formula calculates different values of theta_i for each codeword based on its index i, creating a set of unique codewords in the codebook.
5. The method of claim 1 , wherein n is equal to one, and wherein the rank 1 codebook is generated using generator codeword index codeword u i i C K,N t i (1, :) i ε [0, . . . , (2 K − 1)].
The wireless communication method where rank n equals one constructs the rank 1 codebook using a generator codeword index, codeword ui, with index i, producing Ci based on K and Nt. The indexing method ensures that for each index i within the range [0 to (2^K - 1)], a corresponding codeword ui is selected.
6. The method of claim 1 , wherein n is equal to one, and wherein the i-th codeword of the rank 1 codebook is a first column of the transformed base matrix.
In the wireless communication method where the rank n is equal to one, the i-th codeword of the rank 1 codebook is simply the first column extracted from the transformed base matrix. This greatly simplifies the codebook generation, making the first column representative of the whole codeword.
8. The method of claim 7 , wherein N t =8, K=6, and the rank 2 codebook is generated using generator codeword index codeword u i 4i C K,N t i (1:2, :)/{square root over (2)} i ε [0, . . . , 15] 4i + 1 C K,N t i (1:4, :)/{square root over (2)} 4i + 2 C K,N t i (1:6, :)/{square root over (2)} 4i + 3 C K,N t i (1:8, :)/{square root over (2)}.
A wireless communication method with N_t (number of transmit antennas at the controller) = 8 and K (number of bits in rank n codebook) = 6. The rank 2 codebook is generated using a generator codeword: codeword u_i, at index 4i. The codewords used are: C (K,N_t) at index i (columns 1 and 2) divided by sqrt(2), i is between 0 and 15; C (K,N_t) at index (4i + 1) (columns 1 to 4) divided by sqrt(2); C (K,N_t) at index (4i + 2) (columns 1 to 6) divided by sqrt(2); C (K,N_t) at index (4i + 3) (columns 1 to 8) divided by sqrt(2).
9. The method of claim 1 , wherein n is equal to two, and wherein the i-th codeword of the rank 2 codebook comprises two columns of the transformed base matrix.
In the wireless communication method, where the rank "n" is set to two, each codeword within the rank 2 codebook consists of two columns extracted from the transformed base matrix. These two columns are selected based on the codeword's index within the codebook.
10. The method of claim 1 , wherein an element in row p and column q of the DFT N t base matrix is expressible as D F T N t ( p , q ) = 1 N t ⅇ 2 π pq N t j , where p and q are positive integer values less than or equal to N t .
The wireless communication method has the element in row p and column q of the DFTNt base matrix defined as DFTNt(p, q) = (1/Nt) * e^(j * (2 * pi * p * q) / Nt). 'p' and 'q' are positive integers, less than or equal to Nt. This formula specifies the construction of the base matrix used to derive the codebook.
12. A method for controller operation, the method comprising: receiving, by the controller, an index from a communications device, the index corresponding to a codeword in a rank n codebook, and the codeword representing a quantized channel estimate, where n is a non-negative integer value representing a number of transmit antennas at a controller; computing a precoding matrix from the codeword; precoding data to be transmitted to the communications device in accordance with the precoding matrix; and transmitting the precoded data to the communications device, wherein the rank n codebook comprises 2 K codewords with each codeword comprising n columns of a transformed base matrix of dimension N t ×N t , where K is a positive integer number representing a number of bits in the rank n codebook, and N t is a number of transmit antennas at the controller, and wherein an i-th codeword of the rank n codebook is expressible as C i =diag(u i )×DFT N t , where DFT N t is the base matrix, u i (n)=e jπ sin(k sin(θ t ))n , iε[0, . . . , 2 K −1], nε[0, . . . , N t −1], k is a constant, and θ i ε[0,2π], and diag( ) is a function that returns diagonal elements of u i .
A base station in a wireless communication system receives an index from a mobile device. This index corresponds to a codeword in a "rank n" codebook, which represents a quantized estimate of the channel between the base station and the mobile device. The base station then computes a precoding matrix based on the received codeword. The base station precodes data intended for the mobile device using this precoding matrix and transmits the precoded data. The i-th codeword (Ci) is calculated as diag(ui) multiplied by DFTNt, where ui(n) = e^(j * pi * sin(k * sin(theta_i)) * n). DFTNt is the base matrix.
13. The method of claim 12 , further comprising transmitting a pilot.
The base station operation of receiving an index from a mobile device, calculating a precoding matrix, precoding data, and transmitting it also includes the step of transmitting a pilot signal. This pilot signal is used by the mobile device to estimate the channel characteristics.
14. The method of claim 12 , wherein n is equal to one, and wherein k = 2 3 π 9 , θ i = - π 3 + 2 i π 45 , i ∈ [ 0 , … , 2 K - 1 ] , and n ∈ [ 0 , … , N t - 1 ] .
The base station operation of receiving an index from a mobile device, calculating a precoding matrix, precoding data, and transmitting it also has the following specifics: Rank n is equal to one. The constant k is set to (2/3)*(pi/9), and theta_i = -pi/3 + 2 * i * pi/45, where i ranges from 0 to (2^K - 1), and n ranges from 0 to (Nt - 1).
15. The method of claim 12 , wherein n is equal to two and N t =8, K=6, and the rank 2 codebook is generated using generator codeword index codeword u i 4i C K,N t i (1:2, :)/{square root over (2)} i ε [0, . . . , 15] 4i + 1 C K,N t i (1:4, :)/{square root over (2)} 4i + 2 C K,N t i (1:6, :)/{square root over (2)} 4i + 3 C K,N t i (1:8, :)/{square root over (2)}.
A base station operates where Rank n is two, Nt=8 (transmit antennas), K=6 (codebook bits). The rank 2 codebook uses: codeword ui, at index 4i. The codewords used are: C (K,N_t) at index i (columns 1 and 2) divided by sqrt(2), i is between 0 and 15; C (K,N_t) at index (4i + 1) (columns 1 to 4) divided by sqrt(2); C (K,N_t) at index (4i + 2) (columns 1 to 6) divided by sqrt(2); C (K,N_t) at index (4i + 3) (columns 1 to 8) divided by sqrt(2).
16. A communications device comprising: a channel estimate unit coupled to a signal input, the channel estimate unit configured to estimate a channel between the communications device and a controller serving the communications device; a quantization unit coupled to the channel estimate unit, the quantization unit configured to quantize the channel estimate in accordance with a rank n codebook and to output an index to a codeword from the rank n codebook in accordance with a codeword selection mechanism; a codebook store coupled to the quantization unit, the codebook store configured to store the rank n codebook; and a transmitter coupled to the quantization unit, the transmitter configured to transmit the index to the controller, wherein the rank n codebook comprises 2 K codewords with each codeword comprising n columns of a transformed base matrix of dimension N t ×N t , where K is a positive integer number representing a number of bits in the rank n codebook, and N t is a number of transmit antennas at the controller, and wherein an i-th codeword of the rank n codebook is expressible as C i =diag(u i )×DFT N t , where DFT N t is the base matrix, u i (n)=e jπ sin(θ t ))n , iε[0, . . . , 2 K −1],nε[0, . . . , N t −1], k is a constant, θ i ε[0,2π], and diag( ) is a function that returns diagonal elements of u i .
A mobile communications device estimates the channel between itself and a base station. A quantization unit quantizes this channel estimate using a "rank n" codebook, selecting a codeword based on a predefined mechanism. The codebook is stored in a codebook store. The device transmits the index of this selected codeword to the base station. The i-th codeword (Ci) is calculated as diag(ui) multiplied by DFTNt, where ui(n) = e^(j * pi * sin(k * sin(theta_i)) * n). DFTNt is the base matrix.
17. The communications device of claim 16 , wherein the channel estimate unit is configured to estimate the channel in accordance with a measurement of a pilot transmitted on the channel.
The mobile device, with its channel estimation, quantization and transmission functionality, estimates the channel by measuring a pilot signal transmitted by the base station. The characteristics of the pilot signal are used to derive the channel estimate.
18. The communications device of claim 16 , wherein n is equal to one, and wherein the i-th codeword of the rank 1 codebook is a first column of the transformed base matrix.
The mobile device, with its channel estimation, quantization, and transmission functionality, operates with the rank 'n' set to 1, the i-th codeword of the rank 1 codebook is the first column of the transformed base matrix. This uses a simplified method for generating codewords for rank 1 transmissions.
19. The communications device of claim 16 , wherein n is equal to two, and wherein the i-th codeword of the rank 2 codebook comprises two columns of the transformed base matrix.
The mobile device, with its channel estimation, quantization and transmission functionality operates with rank n is equal to two, and the i-th codeword of the rank 2 codebook comprises two columns of the transformed base matrix. This configuration defines the structure and size of the codebook used for quantizing channel state information.
20. A controller comprising: a feedback decoder coupled to a signal input, the feedback decoder configured to decode feedback information from a communications device to produce an index, where the index corresponds to a codeword in a rank n codebook, the codeword corresponding to a quantized version of channel state information for a communications channel between the controller and the communications device; a beamforming vector compute unit coupled to the feedback decoder, the beamforming vector compute unit configured to compute a beamforming vector from the codeword; a transmit beamforming unit coupled to the beamforming vector compute unit and to a data input, the transmit beamforming unit configured to apply the beamforming vector to data provided by the data input; and a transmitter coupled to the transmit beamforming unit and to an antenna, the transmitter configured to transmit beamformed data provided by the transmit beamforming unit to a source of the feedback information using the antennas, wherein the rank n codebook comprises 2 K codewords with each codeword comprising n columns of a transformed base matrix of dimension N t ×N t , where K is a positive integer number representing a number of bits in the rank n codebook, and N t is a number of transmit antennas at the controller, and wherein an i-th codeword of the rank n codebook is expressible as C i =diag(u i )×DFT N t , where DFT N t is the base matrix, u i (n)=e jπ sin(k sin(θ t ))n , iε[0, . . . , 2 K −1], nε[0, . . . , N t −1], k is a constant, θ i ε[0,2π], and diag( ) is a function that returns diagonal elements of u i .
A base station controller decodes feedback information received from a mobile device to obtain an index. This index corresponds to a codeword within a "rank n" codebook, representing a quantized version of the channel state between the controller and the mobile device. A beamforming vector is then calculated based on this codeword. This vector is applied to data being transmitted to the mobile device, and finally, the beamformed data is transmitted. The i-th codeword (Ci) is calculated as diag(ui) multiplied by DFTNt, where ui(n) = e^(j * pi * sin(k * sin(theta_i)) * n). DFTNt is the base matrix.
21. The controller of claim 20 , wherein the transmitter is further configured to transmit a pilot.
The base station controller operation of decoding feedback, computing a beamforming vector, applying it to data, and transmitting the data, also includes the transmission of a pilot signal. This pilot assists the mobile device in estimating the channel state.
22. The controller of claim 20 , wherein n is equal to one, and wherein k = 2 3 π 9 , θ i = - π 3 + 2 i π 45 , i ∈ [ 0 , … , 2 K - 1 ] , and n ∈ [ 0 , … , N t - 1 ] .
In the base station controller functionality of decoding feedback, computing a beamforming vector, precoding data and transmitting, rank n is equal to one. The constant k is set to (2/3)*(pi/9), and theta_i = -pi/3 + 2 * i * pi/45, where i ranges from 0 to (2^K - 1), and n ranges from 0 to (Nt - 1).
23. The controller of claim 20 , wherein n is equal to two and N t =8, K=6, and the rank 2 codebook is generated using generator codeword index codeword u i 4i C K,N, i (1:2,:)/{square root over (2)} iε[0, . . . , 15] 4i + 1 C K,N, i (1:4,:)/{square root over (2)} 4i + 2 C K,N, i (1:6,:)/{square root over (2)} 4i + 3 C K,N, i (1:8,:)/{square root over (2)}.
A base station controller operates where Rank n is two, Nt=8 (transmit antennas), K=6 (codebook bits). The rank 2 codebook is generated using a generator codeword: codeword ui, at index 4i. The codewords used are: C (K,N,t) at index i (columns 1 and 2) divided by sqrt(2), i is between 0 and 15; C (K,N,t) at index (4i + 1) (columns 1 to 4) divided by sqrt(2); C (K,N,t) at index (4i + 2) (columns 1 to 6) divided by sqrt(2); C (K,N,t) at index (4i + 3) (columns 1 to 8) divided by sqrt(2).
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
March 23, 2010
July 2, 2013
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.